Real-time procedural audio generation—no pre-recorded looping audio files.
Plays a 3-second logarithmic sine sweep to measure transfer function, $T_{60}$, and notch out resonant standing waves.
Configure adaptive operating profiles and sharp psychoacoustic notch therapy filters.
Carves a 48 dB/oct band-stop notch into synthesized masking noise at the tinnitus resonance.
Speech band ($500\text{ Hz – }4\text{ kHz}$) priority masking to eliminate semantic eavesdropping.
Strict $<34\text{ dBA}$ baseline with high-kurtosis transient blunting for nighttime sleep.
Vocal de-escalation logic that dims output to cue occupants to speak quietly.
IndexedDB continuous 1 Hz time-series data visualization & compliance export.
Executes rigorous unit and integration test assertions across all engine modules.
All 5 Sprints fully realized in production code (v1.0.1-GA).
| Feature Module | Sprint | Implementation Target | Status |
|---|---|---|---|
| 24-Bark ISO 532-1 & DIN 45692 Engine | Sprint 1 | Specific loudness $N'(z)$ & envelope AM demodulation | Complete (v1.0.1-GA) |
| Procedural Biophilic Soundscapes | Sprint 2 | Rain, water, alpine wind & habituation drift | Complete (v1.0.1-GA) |
| Room Mode PEQ & Chirp Calibration | Sprint 3 | Transfer function analysis & 12-band notch biquads | Complete (v1.0.1-GA) |
| Smart Context & Tinnitus Notch | Sprint 4 | Sleep cushion, Deep Focus & 1–14 kHz notch filter | Complete (v1.0.1-GA) |
| 24h Heatmaps & Compliance Exporter | Sprint 5 | IndexedDB time-series, report generator & QA suite | Complete (v1.0.1-GA) |